kta-prd-pipeline

Decompose a validated idea brief into per-feature PRDs with nested screens and a navigation graph.

Updated May 31, 2026
One-click install
npx skills add https://github.com/decoutkhanqindev/Lich-Viet-Loc-Phat --skill kta-prd-pipeline-decoutkhanqindev
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: kta-prd-pipeline
Source: https://github.com/decoutkhanqindev/Lich-Viet-Loc-Phat/tree/main/.claude/skills/kta-prd-pipeline
Command: npx skills add https://github.com/decoutkhanqindev/Lich-Viet-Loc-Phat --skill kta-prd-pipeline-decoutkhanqindev

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve? Turning a validated app idea into structured, per-feature Product Requirements Documents is slow and inconsistent when done by hand. This Skill automates the decomposition of an idea brief into 4–8 feature PRDs with nested screen definitions, a Mermaid navigation graph, and a master index. ## Core Features & Use Cases - Feature Decomposition: Reads idea-brief.md feature seeds and proposes 4–8 features grouped into five fixed user-journey buckets (Acquisition, Activation, Core Loop, Retention, Settings), with a hard cap of 8 and monetization explicitly out of scope. - Per-Feature PRD Generation: Writes one markdown PRD per feature using a strict template covering user stories, screens, functional and non-functional requirements, data needs, analytics events, and out-of-scope items, with sensible pre-filled defaults. - Cross-Feature Review: Generates a Mermaid navigation graph, detects shared components and naming conflicts, and writes a master INDEX.md. - Use Case: After validating an app idea with the idea pipeline, run this Skill to produce a complete plans/{slug}/prd/ directory of feature PRDs ready to hand off to a design-spec stage. ## Quick Start Ask the AI to decompose the validated idea brief in plans/{slug}/idea-brief.md into per-feature PRDs with screen lists and a navigation graph.

Frequently Asked Questions about kta-prd-pipeline

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I turn an app idea brief into feature PRDs?▼

Run this Skill after your idea-brief.md is validated with 5–8 feature seeds. It proposes 4–8 features grouped by user-journey buckets, then writes one PRD markdown file per feature under plans/{slug}/prd/ with screens, requirements, and analytics events.

What inputs does the PRD pipeline require?▼

It requires plans/{slug}/idea-brief.md containing a feature_seeds section with 5–8 entries, typically produced by the kta-idea-pipeline stage. If the file or seeds are missing, the Skill stops and prints a clear error telling you to run the idea pipeline first.

Can I include paywall or subscription features in the PRDs?▼

No. Monetization features such as paywalls, IAP, subscriptions, billing, and ads are explicitly out of scope in this phase. The Skill drops those seeds and tells you to defer them to a later phase focused on monetization.

Is there a limit on how many features the PRD pipeline generates?▼

Yes, there is a hard cap of 8 features. If you try to add a ninth, the Skill refuses and asks you to drop an existing feature, consolidate two adjacent features, or defer the new one to a V2 list.

What happens after all the PRD files are written?▼

After writing INDEX.md and nav-graph.mmd, the Skill runs a mandatory handoff prompt offering three choices: auto-chain to the kta-design-spec skill for all PRDs, run design for one selected feature PRD, or exit so you can read the PRDs first.